Storage Controller Pooling for AI Data Access and CPU Offload

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing AI systems face challenges in managing complex storage interfaces and optimizing data storage across different non-volatile storage devices, leading to inefficiencies in performance and cost.

Innovation Solution

A storage controller is introduced that discovers storage devices, creates pools based on performance, and manages address spaces, enabling efficient data transactions and pool management through encapsulation, reducing the need for direct software management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If software directly manages storage interfaces and device pooling, then fine-grained control over storage devices is achieved, but system complexity and software resource consumption increase

Engineering Contradiction:

Improvestorage management complexityVSAvoidsoftware complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent introduces a storage controller as an intermediary device between the software and storage devices. The storage controller autonomously performs device discovery, pooling, and address space management, eliminating the need for software to directly manage these complex operations. This mediator approach resolves the contradiction by shifting management complexity from software to a dedicated hardware controller.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The storage controller is designed to autonomously perform storage management functions without requiring software intervention. It automatically discovers storage devices, creates pools based on device characteristics, and manages address spaces independently. This self-service capability reduces software complexity while maintaining operational control.

Inventive Principle:

Principle #25Self-service

2Productivity

If software manages multiple storage interfaces directly, then detailed control over data placement is achieved, but CPU cycles and memory resources are consumed

Engineering Contradiction:

Improvedata access efficiencyVSAvoidCPU and memory resource consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The storage controller acts as an intermediary that handles resource-intensive storage management operations in hardware, freeing CPU and memory resources. It manages device pooling, address space allocation, and data placement autonomously, enabling efficient data access without consuming software resources.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces software-based storage management with a hardware-based storage controller. This substitution moves management functions from the software/CPU domain to a dedicated hardware device, reducing CPU cycles and memory usage while maintaining or improving data access efficiency through specialized hardware optimization.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Speed

If volatile storage is used as primary storage for AI systems, then fast data access is achieved, but cost and capacity are limited

Engineering Contradiction:

Improvedata access speedVSAvoidstorage capacity and cost

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The patent segments storage into performance-based pools that can include both volatile and non-volatile devices. Fast volatile storage devices are grouped into high-performance pools for speed-critical operations, while non-volatile devices form capacity pools for cost-effective storage. This segmentation allows the system to optimize between speed and capacity/cost for different data workloads.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different pools are created with distinct performance characteristics matching specific access needs. High-performance pools use fast volatile storage for data requiring rapid access, while capacity pools use slower non-volatile storage for less time-sensitive data. This local quality differentiation allows the system to achieve fast access where needed while maintaining overall cost-effectiveness and capacity.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20260079638A1Data object-based storage controllers

Publication Date: 2026.03.19 AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE LTD

AI summary

An example computing system includes a hardware platform including a central processing unit (CPU), volatile memory, storage devices, and a storage controller; software, executing on the hardware platform, the software configured to: discover, by the storage controller, the storage devices; create, by the storage controller, pools of the storage devices; obtain, by first software executing on the CPU, information describing the pools; determine, by the first software in cooperation with the storage controller, an address space for each of the pools; and communicating, from the software to the storage controller, a storage transaction referencing a first pool of the pools and a first address in the address space for the first pool.